INDEX

Explanations

adjectives and verbs related to reflections or self-examination

oai_token-act-pair · gpt-3.5-turbo

references to mirrors and reflective imagery

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stant

-0.79

--------------------------------------------------------

-0.76

cific

-0.73

anooga

-0.72

ensable

-0.72

reek

-0.71

aii

-0.69

mercial

-0.69

estern

-0.67

lished

-0.67

POSITIVE LOGITS

ror

0.97

 neuron

0.89

 shine

0.83

ocular

0.82

 reflection

0.80

 mirror

0.79

angelo

0.79

 Mirror

0.79

 shards

0.76

 image

0.72

Activations Density 0.017%

adjectives and verbs related to reflections or self-examination

references to mirrors and reflective imagery

No Comments

No Known Activations